The permeability of the small intestine was measured and jejunal biopsy performed in 39 children with gastrointestinal disorders. Intestinal permeability was measured using orally administered mannitol and lactulose as probe molecules in an isotonic solution (274 mOs/l), and the results were expressed as the ratio of the urinary excretion of the two sugars over 5 h. Urine samples were analysed for mannitol and lactulose content by high performance liquid chromatography. Children with small intestinal mucosal damage, irrespective of the cause, had a significantly lower (P less than 0.001) mannitol excretion (mean recovery 1.21% of ingested dose) than those with a normal mucosa (mean recovery 5.3%), while lactulose excretion did not differ (P greater than 0.05). The mean value of the lactulose:mannitol urinary excretion ratios was significantly higher (P less than 0.001) in subjects with an abnormal mucosa (0.98) compared to those with a normal mucosa (0.2). Using the mean plus two standard deviations of the normal mucosa group to define the upper limit of normal, all lactulose: mannitol excretion ratios from the abnormal mucosa group were above this limit. The results of this study show that the sugar permeability test is a sensitive, non-invasive screening test for jejunal mucosal damage in children and shows good correlation with jejunal biopsy results.